An interview with catastrophe insurance expert Robert Muir-Wood
September 27th 2020. Melissa Sterry
With over twenty years of experience developing probabilistic catastrophe models, Dr. Robert Muir-Wood works to enhance approaches to natural catastrophe modelling. The lead author of the 2007 IPCC 4th Assessment Report and 2011 IPCC Special Report on Extremes, he holds a PhD in Earth Sciences from University of Cambridge and is vice-chair of the OECD panel on the Financial Consequences of Large-Scale Catastrophes. Here Robert shares his thoughts on the past, present, and possible future of catastrophe insurance…
MS: As you discussed in ‘The Cure for Catastrophe: how we can stop manufacturing natural disasters’ , colonialism led to architectural concepts fit for one region, spreading to others for which they were not. What are the worst-case historical examples you’ve come across?
RMW: It is a feature of colonialism to assume that importing the latest architectural style from the motherland will help unify the empire and manifest progress. The British exported brick house-construction to earthquake prone locations, where such buildings proved deadly, whether this was the ‘little London’ that had been constructed in Port Royal, Jamaica before 1692, or the factories built in 1880s Japan. Britain was not unique in this regard: in the 18th century we had the dangerous heavy stone baroque architecture employed around the Spanish colonies of Central and South America.
MS: History repeating itself, what are the worst-case ‘one-size-fits’ all approaches you’ve seen of this past couple of decades?
RMW: As I explored in ‘The Cure for Catastrophe’, if you build a flood defence people will come and live behind it, assuming they have absolute protection. In New Orleans in 2005, faced with the storm surge from Hurricane Katrina, flood walls collapsed before being overtopped. The walls were later discovered to be 40cm lower than intended because the whole city, including the survey benchmarks, was rapidly sinking. Among the people who lived below sea-level, more than a thousand died, unable to escape the floodwater in their single storey houses. In 2011 we saw an even greater tragedy in Japan where 20,000 were killed in the tsunami that followed (15-20 minutes after) the Magnitude 9 earthquake along the Pacific coast. After feeling the shaking, many of those who lived in the coastal floodplain made no attempt to evacuate uphill from their houses because they assumed the government-built tsunami walls would protect them. In some towns the tsunami was twice as high as the walls. In Chile they don’t build tsunami walls, so those who live at the coast know, after feeling a strong earthquake, they will need to run inland. Also, without any protection people prefer not to live in the coastal zone. The death toll of the daytime 2011 tsunami in Japan was around a hundred times higher than in the night-time tsunami from the similar sized 2010 earthquake in Chile.
People can only see the wisdom of investing in raised defences, after a flood. The height of a flood defence is based on assessing the benefits (the annualized cost of the floods that don’t happen) relative to construction costs. The hazard may have risen since the flood defences were built, especially where rising sea-levels, higher rainfall or sinking land are raising the chance of extreme water levels. Or the original understanding of the flood hazard may have been incomplete.
MS: In what other ways do you think humanity is building in the wrong way at the wrong place, and why?
RMW: We need to have a full appreciation of the hazard climate over the expected lifetime of a building to know what protections should be included, or whether a structure should be built at this location at all.
There is a universal tragedy of informal settlements that develop on land in a city considered too hazardous (too steep, too landslide-ridden, too flood-prone) for official licensed development. In this situation, which exists in many developing world cities, the informal settlement, (the barrios) situates itself in the most dangerous location of all, because, without any official development, this space is unoccupied. The most dangerous land then gets built on at the highest density and without any supervision. This led to more than 100,000 fatalities in the 2010 earthquake in Haiti. When Cat 5 Hurricane Dorian hit the Bahamian islands of Abaco and Grand Bahama in 2019 [pictured below], the accompanying storm surge, caused uncounted fatalities among informal workers from Haiti, living on land too close to sea-level for any formal development. Without any title to the land on which they live, new migrants will refuse to leave their homes, as the flood bears down on them.
Throughout the Caribbean, development is set back, by each major disaster. And yet if it was only possible to raise the standard of construction and to prevent development in the path of floods and landslides, it would be possible to develop the resilience that enables a city and an economy to rebound from any hazard.
MS: Few, if any industries are as adept at assessing risk as the insurance sector. What are the primary lessons in relation to natural – and not so natural - hazards that architects, planners, and policymakers could learn from insurers?
RMW: Insurers are good at evaluating the risk of what they insure. Inevitably they hold less experience at evaluating the risks of what is not insured, whether because it is covered by a government scheme or is an informal settlement.
Insurers should be encouraged to expand their remit, and to be engaged at the initiation of any new building project. Will this building be insurable over its lifetime? At present, whether it is in the wildlands of California, or the beach front in Australia, insurers are expected to sustain coverage, independent of how the risk is changing. Insurance contracts are only for a year, but society has expectations that insurance will be sustained for a lifetime. In a world in which risk is changing, many problems emerge from this clash of time horizons.
MS: The Great Fire of London catalysed the catastrophe insurance industry. How do you see fire shaping the insurance industry in the coming decade, and why?
RMW: Fire is the original universal insurance coverage. The minimum insurance coverage for a property covers fire. And yet fire is also the peril that shows the most notable higher order increase in a warmer world. Forests, no longer able to tolerate the climate, die and become the fuel for fires. More than a hundred million dead trees in California are the legacy of the 2011-2018 droughts. Warming can lead to longer summers and more fire starts, as well as baking vegetation into fuel. Warming can also generate more intense rainfalls in a shorter winter leading to more vegetation. Burn back the forest and the climate gets even hotter. And all this accompanies the great American desire to live close to nature in the wildlands: to live under the shade of a benevolent tree that one day may set your house on fire. Even more than flood, wildfire is going to present the first big challenge of climate change to insurance.
MS: I’ve posited the possibility that insurers may migrate away from a model that compensates for loss of property to fire with cash settlements, and towards a system which provides homeowners with access to the data, suppliers, and expertise they would need to rebuild their homes. In this hands-on recovery scenario, the insurer would play an active role in enabling their customers to rebuild with resilience. What are your thoughts on this proposal?
RMW: A key question surrounds how to ensure a destroyed building is rebuilt to the latest building code. Insurers might invest in resilience if the insurance contract lasted for ten years when they would have a sustained financial interest in risk reduction. For a one year contract, there seems no advantage in their investing.
Insurers are pressured by regulators to settle claims promptly. However the owner of a destroyed property might want to take a different approach to reconstruction, and to have part of the compensation redirected to providing the professional expertise from an architect and builder, to create a property different, and more resilient, to that which had been destroyed. This could include investing the owner’s own money, or extending their borrowing. An insurer could define a coverage that involved providing professional support as part of total compensation after a 100% loss. However one has to be careful to avoid moral hazard, when the owner is motivated to see their current building destroyed.
MS: Myriad classes of sensing, processing, and predictive technologies now enabling real and near-to-real time monitoring of multiple Earth systems and the natural hazards they make manifest, do you see insurers developing tools to help their customers anticipate and prepare for likely near-future wildfires, floods, storms, droughts, heatwaves, and pandemics?
RMW: Anticipatory information is already provided in advance of hurricanes, but clearly the provision of such data could be taken much further. Warnings need to be given in sufficient time that action could be taken. Information also needs to be delivered with probabilities, to reflect the inherent uncertainty of forecasts. There is always the danger of ‘crying wolf’ after which warnings become ignored.
MS: The insurance sector has consistently produced some of the foremost considered and informative foresight reports with respect to climate change and its possible consequences. Which do you recommend to readers?
RMW: In some countries already worst impacted by climate change the insurance industry has taken a lead in developing objective information on how climate change is expected to affect catastrophe risk. Australia is something of a leader in this regard, and in work commissioned and presented by the insurer Insurance Australia Group: IAG - the largest general insurer in Australia and New Zealand. Read here.
For many years, global reinsurers like Munich Re and Swiss Re have produced open reports demonstrating the long term rise in insurance costs and promoting action on climate change. Read here, here, and here.
MS: Collaboration is central to solving the complex problems humanity faces, how do you think the insurance sector could better share its knowledge, its skills, and its wider capacity to help bring about best-case, not worst-case environmental outcomes?
RMW: The insurance sector is expected to sustain coverage in a world in which risk is rising. Insurers need to rise to the challenge as leaders in informing society about risk, and in championing the actions that will be required to maintain insurability.
MS: Do you see a role for insurers in helping to educate students of architecture, urban design, planning, and policy on the risks of climate change, and particularly as relates to fire in both its wild and urban context?
RMW: The risk professionals to be found inside the largest insurers and reinsurers, as well as the supporting risk modelling consulting sector, could, and should, be better harnessed to help educate students. Insurers gather data around claims and loss, which is often unavailable to architects and planners. Architects focus on how to build, not what brings about damage and demolition. We also need to teach a holistic perspective on risk. How risk relates to poverty. How a flood defence does not eliminate risk, and can even increase risk, when it encourages new development. How not building on dangerous land in a city, can raise risk, when it encourages informal settlements. How the geography of risk is shifting. The whole concept of Risk should be taught to children at school. The game of Snakes and Ladders is a good introduction to the randomness, and consequences of risk. Students of relevant disciplines should be taught how risk is measured and mapped, and how actions related to risk have consequences.
MS: The year is 2030. What’s the worst-case scenario you visualise?
RMW: Unfortunately only disasters, both acute and chronic, are going to motivate action to reduce greenhouse gas emissions. So the worst case scenario in terms of disasters linked to climate change, could provide the strongest arguments for driving political action. But, after a crisis there is always a danger of returning to the preceding state, and making no permanent change. There are far too many vested interests, and lobbyists, intent on preserving the emissions status quo. And then there is the challenge that the most adverse outcomes are in the future.
I am most concerned about the loss of biodiversity, from which there is no way back. Climate change will combine with the impacts of agricultural expansion and population increase to destroy natural habitats.
MS: The year is 2030. What’s the best case scenario you visualise?
RMW: One can dream about a world in which there is co-ordinated action, in which people are motivated to give up those activities, like flying or driving, they learnt to live without in the Covid-19 lockdown. The insurance industry needs to become motivated to drive societal change, not cover the risks produced by maladaptation.
Find out more about Robert’s thoughts on mitigating catastrophe here and here.
Images: [Top] Bobcat Fire, El Dorado Fire, and Apple Fire, September 15th 2020; [Middle] Parts of the Abaco Islands after Hurricane Dorian, September 5th 2019, and [Bottom] Oklahoma Fire Complex, April 17th 2018, all containing modified Copernicus Sentinel data and processed by Pierre Markuse.